---
id: "lib-2026-10-29-tokwatt-an-open-boundary-annotated-database-of-ai"
type: "library"
title: "TokWatt: an open, boundary-annotated database of AI inference energy, from chat answers to agents"
lang: "EN"
org: "arXiv"
source_type: "Paper"
date: "2026-10-29"
week: "2026-W44"
url: "https://nosov.ai/library/lib-2026-10-29-tokwatt-an-open-boundary-annotated-database-of-ai/"
original_url: "https://arxiv.org/abs/2610.39846"
tags: ["ai technology", "energy", "cost"]
---

# TokWatt: an open, boundary-annotated database of AI inference energy, from chat answers to agents

## English summary

134 estimates from 13 sources harmonised: 0.34 Wh short chat answer, 4.7 Wh reasoning answer, 150 Wh per coding-agent task; 5 production chat estimates agree within 1.8x.

Original: https://arxiv.org/abs/2610.39846